Testing AI in Daily Work: What Worked and What Didn't

What Actually Worked

The part that surprised me the most was meeting summaries, by far.

How I Used to Do It

1. I pasted loose notes I took during the call itself

2. I asked for an organized summary in short bullet points

3. I separated what became a decision from what remained open for later

The points that mattered most in daily work were these three:

• Meeting summaries in bullet points, always in the same format

• Drafting awkward emails, like polite follow-ups or apology for delays

• Reviewing text I had already written myself, just to polish it

This alone saved about twenty minutes of my late afternoon, every single day of the week, with no exception. It doesn't sound like much, but over two weeks that turns into almost four hours given back to my time.

What Didn't Work

I tried using AI to prioritize my daily tasks, just typing tell me what to do first and that's it. It didn't work well, because true priority depends on context I never write down anywhere — informal agreements with the boss, urgency that only I know exists at that moment.

The tool had no way of guessing any of that, and the suggestions were always too generic to be truly useful in daily work.

I also tried writing a longer report from scratch directly with the tool, and the result came out correct in content, but without my voice in it — it felt like a report written by someone else, not me.

Is It Worth It?

After two weeks, my conclusion is that it is worth using AI for specific and well-defined tasks — summarizing, drafting, reviewing — but not for decisions that depend on context that only exists inside your head. In the end, I used the saved time for things that actually required my real attention.